extractClassifications

import pandas as pd
import numpy as np

pokemonTable = pd.read_csv("dataClean/Pokemon.csv")

# Get a list of all unique values in the classification-column and sort it alphabetically
classificationList = pokemonTable.classification.unique()
classificationList = np.sort(classificationList)

# Build a dictionary from the list, mapping the name to the corresponding index
classificationDict = {name: index for index, name in enumerate(classificationList)}

# Translate all names into numerical values using the created dictionary
pokemonTable.classification = pokemonTable.classification.map(lambda x: classificationDict[x])

# Create new table with all index <-> name mappings
classificationTable = pd.DataFrame(classificationDict.keys(), columns=["name"])
# Cleanup name of row 331 because it contains a weird value ._.
classificationTable.name.iloc[331] = "Mischief Pokémon"

# Replace every é with e because special character encoding in mySQL is hell ._.
classificationTable.name.replace("é", "e", regex=True, inplace=True)

classificationTable.to_csv("dataClean/Classifications.csv", encoding="utf-8")
pokemonTable.to_csv("dataClean/Pokemon.csv")